Difficulty in emotional perception is a key trans-diagnostic impairment. Emotion perception deficits contribute to mental health treatment costs and incur significant social and economic costs, more broadly. Inability to perceive emotion can have a profound impact on professional and personal realms, due to losses in worker productivity and disruptions on social functioning. Yet scientists lack of mechanistic understanding of emotion perception and deficits therein. This represents a critical barrier to treatment and prevention within this major facet in social functioning. I propose that surmounting this critical barrier can be achieved by examining emotion perception through a cross-cultural lens, using a process-based theoretical approach. Cultural comparisons can provide a lens on normative variation that is typically obscured within a single cultural context. The proposed research aims to reveal the ways in which the underlying mechanisms that contribute to emotion perception are contextually bound and culturally variable. I will examine two processes: mentalizing (i.e., how much people make inferences about the inner feelings) and action perception (i.e., how much people identify situated behaviors). It has long been assumed that health is universally defined by the engagement of mentalizing during emotion perception. Yet my prior research identified a culture (the Himba of Namibia) where this is not the case. I plan to continue study individuals from the Himba cultural context, and expand to a Chinese cultural context, to further probe cultural variation in the mechanisms of normal emotion perception. With aim 1, I will better characterize cultural variation in mentalizing and action identification through analysis of natural language use during emotion perception and mentalizing tasks. With aim 2, I will investigate the neural underpinnings of cultural variation in mentalizing using ambulatory neuroimaging by examining the functioning of one region that anchors the network underpinning mental state inference the dorsomedial cortex (dmPFC). I predict that people in cultural contexts in which mental state language use occurs more frequently (Aim 1), the dmPFC will be more robustly activated, during both domain specific (emotion) and domain general (mentalizing) tasks (Aim 2). By examining emotion perception as a complex act of social cognition within a cultural framework, I aim to reveal the underlying mechanisms of emotion perception that are consistent across cultural contexts and those that differ. This will allow me t build a more mechanistic model of emotion perception that can inform our understanding of how these processes can go awry to produce disordered emotion perception. Based on this research, future interventions could be tailored based on a mechanistic account. In the United States, treatment for individuals high in action identification (e.g., individuals with Autism) coud be informed by understanding transactions that primarily rely on action identification in other cultures.